Work Packages

Work Package One: Understanding Support Needs

The risk of a suicide attempt is increased in individuals with depressive symptoms. Additionally, the risk of suicidal ideation is higher in youths and young adults with T1D. Depression greatly increases the risk of severe hypoglycemic and hyperglycemic events, particularly in the first six months to one year after diagnosis. People with T1D and persistent elevated symptoms of depression are also unfortunately more likely to experience DKA.

 

Adolescents and young adults with T1D and depression experience poor metabolic control and high rates of hospital admissions as well as longer inpatient stays. Little is known, however, about the true rate of intentional self-injury (ISI) and suicide in those with T1D as it is likely under-estimated due to shortcomings in detection mechanisms of such acts across healthcare.  What is known, however, is that adolescents aged 15-19 years are most likely to present to emergency departments with self-inflicted injuries (9.6 per 1000 visits), self-harm being one of the strongest risk factors for suicide.  In a national cohort, particularly female adolescents with T1D reported highest suicidal behaviors. Therefore, all stakeholders, providers, and people with diabetes must engage to fully assess unmet needs and preferences for how best to address them.  

Goals and Approaches

  • Survey people with diabetes (through online questionnaire) to better understand the extent and prevalence of self-reported ISI and suicidal ideation; obtain personal reported feelings associated with such events and assess what support they believe is needed

  • Survey healthcare professionals to determine awareness of issues associated with ISI and suicide; assess for the availability of resources that provide support and education as well as the ability to access them easily; determine what needs have yet to be met

Work Package Two: Detection of Cases

Rates for cause of death are typically determined by what diagnoses are included on death certificates or by which International Classification of Diseases (ICD) codes are used. Unfortunately, this is an unreliable and imperfect system in general and therefore likely to contribute to the under-reporting of suicide in people with T1D.

 

Goals and Approaches

  • Evidence synthesis: coding of existing ISI and suicidal acts, as well as guidelines and recommendations (where/if they exist) as to how to correctly identify such events and accurately record them in electronic health records

  • Gather recommendations and develop a framework to standardize language used (terminology and codes)

  • Machine learning within large datasets of electronic health records to identify cases of ISI and suicidal acts and identify if patterns/trends exist that may assist with developing screening/assessment tools (see below).

Work Package Three: Identification of risk factors

Identifying psychosocial characteristics that are associated with or are potential predictors of depression, ISI, and suicidal thoughts and behaviors - specifically in those with T1D - can facilitate early detection for those at higher risk as well as aid the development of methods for prevention and intervention.  Advanced treatment technologies in T1D (e.g. CGM, insulin pumps, Bluetooth-enabled insulin pens) yield abundant objective data that can be leveraged to identify risky insulin behaviors and provide ways to measure effectiveness of outcomes for interventions designed to reduce risk-taking behaviors and hopefully improve outcomes in those with depression, ISI, and suicidal ideation/behavior. Utilizing other technological advancements in research, specifically machine learning and neurocognitive investigations, in combination with psychosocial evaluations can lead to the development of multiple modalities that will identify associated factors and predictors of depression, ISI, and suicide.

 

Goals and Approaches

  • Utilize machine learning or other predictive models to allow continuous identification of patterns associated with and predicting ISI/suicide risk using EHR data

  • Identify of new sources of in-clinic patient-reported outcomes data that improve predictive models for ISI/suicide risk

  • Exploration of novel types of data collection (imaging, audio/video, self-management device, physiologic sensor, social media, mobile phone usage, etc.) that may improve predictive models for ISI/suicide risk, including neuro-cognitive aspects of ISI and suicide, psychological assessment, qualitative investigation and assessment of psych risk factors

Work Package Four: Identification and early intervention

Use of diabetes technologies to support self-management has increased rapidly, particularly among the pediatric population. The 2016-17 National Pediatric Diabetes Audit of children and young people in England and Wales reports 32% insulin pump use in this population, an increase from 8% in 2011. The US T1D Exchange Registry (from 2010 to 2018) has also shown an increase in insulin pump use from 57% to 63%. The use of CGM devices also increased in the US during the same timeframe from 7% to 30% respectively. While the use of technology may make the management of T1D less burdensome for some, others often become overwhelmed by the data and some youth feel more stigmatized for having diabetes when wearing CGM. Increased access to libre or CGM can paradoxically increase anxiety around hypo- and hyper-glycaemia and increase diabetes-related burden, however more recent data suggests a decrease or no effect in hypoglycaemia worries. Additionally, because people with T1D require and therefore have ready access to insulin, insulin overdose has been reported as a considered method of use in those with suicidal ideation.

Goals and Approaches

  • Identify incidents of ISI through insulin mismanagement when individuals present to the emergency room (ER) with severe hypoglycemia or DKA

  • Develop a standardized screening approach for patients presenting to the ER with severe hypoglycemia or DKA

  • Make evidence-based interventions (especially access to mental health specialists) immediately available to patients identified to be engaging in ISI.

  • Develop bundle of measures to support tailored onboarding

  • Develop evidence-based guidelines to disseminate best practices